Measuring Peace in a Multi-Ethnic Society Bosnia-Herzegovina 2014
Total Page：16
File Type：pdf, Size：1020Kb
It seems that when human security stems from dissatisfaction with civic life and distrust in institutions, it has a different effect on political integration than when it emerges from satisfaction with personal life. Dissatisfaction with civic governance and distrust in institutions breed insecurity, which translates into a desire to change the system. When insecurity is more closely related to dissatisfaction with personal life, rather than with the state, this leads to less openness to change and the inclusion of the other, or to reconciliation with members of the outgroup. Both of these trends coexist, although possibly not within the same individual, within the Bosniak population. 143 Chapter Five The Bosnia-Herzegovina SCORE: Measuring peace in a multi-ethnic society Maria Ioannou, Nicolas Jarraud, Alexandros Lordos Bosnia-Herzegovina (BiH) was the first country after Cyprus where SCORE was implemented. The main groups studied were Bosniaks, Serbs and Croats. As in Cyprus, the two main dimensions which made up the index in Bosnia-Herzegovina were; social cohesion and reconciliation. The political outcome was political integration defined as readiness for political compromise and the ability to envisage a shared future with other ethnic groups. The presentation of the SCORE results for Bosnia-Herzegovina will follow the same format as for Cyprus, with the results of the descriptive analysis presented first. In this section, we will be presenting the results for all the main dimensions and their indicators, as well as those for other indicators of interest. Where there are significant discrepancies in the demographic break-down they will also be mentioned. This section will be followed by the results of the predictive analysis. This will highlight the indicators of social cohesion and of reconciliation which significantly predict political integration. The chapter will conclude by outlining and discussing the main findings and making corresponding policy recommendations. Methodological highlights The index was calibrated for use in Bosnia-Herzegovina in December 2013 and data was collected in March and April 2014. Participants were citizens of Bosnia-Herzegovina aged 18 and above. Random stratified sampling was used to produce a representative sample of the population. A total of 2,000 respondents were interviewed face-to-face. Of these, 858 were Bosniaks (43%), 847 were Serbs (42%) and 214 Croats (11%). The remaining 144 respondents were from other ethnic groups and were excluded from the analysis. This reduced the sample size to 1,919. The sample comprised 43% male participants and 57% female. The ratio of men to women was about the same across the three ethnic groups. Furthermore, 49% of the sample was recruited from the Federation of Bosnia-Herzegovina (FBiH), 49% from Republika Srpska (RS) and 2% from the neutral, self-governing, Brcko District. Sampling took into account the ethnic composition of each entity; the table below provides a breakdown of each entity’s sample by ethnic group. Total Bosniaks Serbs Croats Entity’s N N (% of entity’s N) N (% of entity’s N) N (% of entity’s N) Federation of Bosnia & Herzegovina (FBiH) 934 698 (75%) 66 (8%) 170 (17%) Republika Srpska (RS) 945 131 (14%) 775 (82%) 39 (4%) Brcko Districs (BD) 40 29 (73%) 6 (14%) 5 (13%) Table 1. Breakdown of each entity’s sample by ethnic group. The majority of participants from the Federation of Bosnia-Herzegovina were Bosniaks, followed by Croats. Serbs formed only a small fraction of the sample. Not surprisingly, Serbs constituted the vast majority of the sample from Republika Srpska. There is also a sizeable Bosniak minority living in RS, which was proportionately represented in the sample. The substantial Bosniak minority resident in RS has made it necessary, where appropriate, for us to disaggregate results from Bosniak respondents by entity (FBiH vs RS). The same was done for Serbs, with the caveat that the results for Serbs living in the FBiH should be interpreted cautiously, since the sample size of 66 was small. This also applies to results from Brcko District. A sample size of 40 is extremely small and makes it impossible for the results from Brcko to be interpreted in isolation. 145 Measuring social cohesion, reconciliation, and political integration As with SCORE Cyprus, our initial task was to identify the indicators of social cohesion and of reconciliation based on the actual data. The process entailed finding a solution that best fitted the data and which did not differ significantly between ethnic groups. Since it is likely that different groups understand abstract concepts, such as social cohesion, differently, in order to carry out a valid comparison between ethnic groups in Bosnia- Herzegovina, we had to verify first that the three groups conceptualised the concepts in the same way. We needed to ensure in other words, that the sub-indicators and indicators defining each dimension did not differ from group to group. Our analyses yielded that the dimension of social cohesion was best defined by three distinct indicators, human security, trust in institutions, and satisfaction with civic life. These indicators differed from those identified as defining social cohesion in Cyprus 2014. In Bosnia-Herzegovina, unlike Cyprus, the representational capacity of institutions and freedom from corruption did not emerge as indicators of social cohesion1. This is because the representational capacity of institutions could not, empirically, (based on our data) be distinguished from trust in institutions. The items measuring trust in institutions and those measuring representation by them were very highly correlated and loaded onto the same factor. We therefore decided to retain the three items measuring trust in institutions in our model, while keeping in mind that trust in institutions and confidence in their representational capacity were essentially the same thing. Human security was measured as one construct in SCORE Bosnia-Herzegovina. In SCORE Cyprus 2014, it was broken down to its constituents, which resulted in measuring three types of security; political, personal, and economic. SCORE Bosnia-Herzegovina was implemented before the break-down of human security was conceptualised, hence its uni-dimensional structure in the index. The figure below shows the three indicators making up the social cohesion dimension and the items through which each of the indicators was measured. 1 The measure of corruption in a new addition to the SCORE Cyprus 2014 which was conducted after SCORE BiH was implemented. We suggest to add this dimension in future reiteration of SCORE BiH. 146 Figure 1. Indicators of social cohesion and items used to measure each indicator. The reconciliation dimension was made up of five indicators; negative stereotypes, intergroup anxiety, social distance, social threats, and positive feelings. These indicators are almost identical to those used in SCORE Cyprus. The only difference being that the indicator ‘active discrimination’ was replaced in this instance by ‘positive feelings’. The items measuring each of these indicators are provided in the figure below. The only indicator that was not measured by multiple items, was positive feelings. To measure positive feelings respondents were asked to rate their feelings towards different outgroups on a scale that ranged from very negative to very positive. Finally, we looked at the items included in the questionnaire measuring Bosnians’ vision for their country. We extracted those items that measured citizens’ views on political integration, which all loaded onto one factor. This factor, which we labelled ‘political integration’, formed the third dimension of the index. The items are presented in the figure below. 147 Figure 2. Indicators of reconciliation and items used to measure each indicator. Figure 3. Items measuring political integration. 148 Results What do the numbers mean? The numbers presented in the descriptive section of the results range from 0 to 10, where 0 indicates very low levels of an indicator / dimension and 10 the maximum possible. The numbers in the predictive section represent regression coefficients which indicate whether and how, indicators relate to each other. Regression coefficients are only presented if they are significant at the 95% level (which means that their p-value is below .05). Greater coefficient values indicate stronger relationships between indicators. A positive value indicates a positive relationship between the two variables, while a negative value indicates a negative one. Part A: Descriptive analysis Social Cohesion and related indicators In this section we will present the results for social cohesion and the indicators which make it up. Results will be disaggregated by entity (FBiH and RS) as well as by ethnic group. Apart from social cohesion, a number of other variables were measured that were theoretically related to it, but which did not, in the end register as predictors of the construct. These variables were; satisfaction with personal life, identity preference and strength of identification with preferred identity, civic engagement and information consumption. Social cohesion Entity level There are discrepancies between entities, with the inhabitants of the Republika Srpska experiencing higher levels of social cohesion than citizens of either the Federation of Bosnia-Herzegovina or the Brčko District (see Table 4). 149 Table 4. Levels of social cohesion at the entity level. We wanted to examine if this discrepancy reflected